---
title: 'Agent Operating Systems: Automate Anything!'
source: 'https://youtube.com/watch?v=QulL6Uqzdmc'
video_id: 'QulL6Uqzdmc'
date: 2026-07-30
duration_sec: 580
---

# Agent Operating Systems: Automate Anything!

> Source: [Agent Operating Systems: Automate Anything!](https://youtube.com/watch?v=QulL6Uqzdmc)

## Summary

This video is a community Q&A session focused on the Agent Operating System (Agent OS), a platform for building and deploying AI agents. The host answers questions about creating custom HR agents, using tools like Pi and Buzz, and comparing popular models like Claude and Hermes.

### Key Points

- **Building a Custom HR Agent** [01:07] — For a large company needing an HR agent that references different data and stores history, the host recommends using a custom GPT for simplicity and shareability, or a tool like Buzz for a more advanced setup.
- **Pi as an AI Agent Framework** [02:42] — Pi is a customizable AI agent framework similar to OpenClaw or Hermes. It's powerful but requires more setup; Hermes is simpler and more ready out of the box. The host notes OpenClaw was built on Pi.
- **Goal Mode in Agent OS** [03:34] — Goal mode is a feature that allows the agent to autonomously work towards a goal, including coding tasks. It's available in both Codeex and Hermes within the Agent OS.
- **Availability of Fable 5 in Claude** [04:15] — Fable 5 is still available in Claude Desktop and is permanently part of the Claude subscription. It's currently better than Opus 5, which is too slow.
- **Hermes vs Claude Usage** [04:44] — According to Open Routers data, Hermes agent is the most used app on the platform, with three times more tokens used than Claude. The host prefers Claude for its Fable 5 model, but acknowledges Hermes' flexibility with free APIs and customization.
- **What is Buzz?** [05:25] — Buzz is described as 'Slack for AI agents'—a user-friendly platform where you can add agents with specialties, operate with subscriptions, and integrate with Hermes. It's beginner-friendly and visual.
- **Member Success: Automated SEO Machine** [06:08] — A new community member built a fully automated SEO machine for his business in his first day using Agent OS, showcasing the platform's rapid deployment capability.
- **Education Technology with Agent OS** [06:38] — The host suggests using the voice agent for language learning or teaching any subject via custom training, and the video manager for creating course videos automatically.
- **Custom Modules in Agent OS** [07:46] — To add custom features, the host recommends describing the desired functionality to the agent, testing, and iterating—a three-step formula using tools like Fable 5 and Nine Router.
- **Token Cost Minimization** [09:08] — For users concerned about token costs, the community provides a token minimization playbook to reduce usage, and the Agent OS can be run for free.

### Conclusion

The Agent OS platform offers a wide range of customizable tools for building AI agents, from simple custom GPTs to advanced setups like Hermes and Buzz, with community support to address user questions and optimize costs.

## Transcript

of the latest questions on the agent operating system. This is a powerful together. We actually added a bunch of new features recently. So for example, that just dropped today. We've added that in. Also got for example free
that in. Also got for example free clawed code with two separate free API routters plugged in which is omniroot and nine router. And then additionally we have a new free AI coder set up here. So you can code over here. You can
view the workspace. You see the history. And you can actually view all your previous creations over here too. And then for example, we have a memory system plugged in. We have for example notebook LM plugged in or Gemini
notebooks. This is now called anti-gravity Gemini 3 uh 5.6 Flash And today we're going to answer some of the latest questions inside our AI profit boardroom community because I know if people inside the community are
setting this up and using it then I know that you probably have similar questions straight into it and we'll start answering some of the latest questions about the agent operating system. So the first one we got from Stephen was about
sub agents and it was basically like how would you consider developing a agent for a large company who want to develop a HR agent right the agent would need to reference different stuff um use it etc. Uh it would be good if it stored the
history as well. Is there a solution for this? Hope that makes sense. Right. So for this setup, I would actually recommend either using a clawed project that you can give members access to, but if you need it to be like much wider,
then you can actually create a custom GPT. So a custom GPT allows you to you can share with other people. And then it's also really easy to use. And that's one of the best things about a custom GPT is like it's sharable. Anyone
can access it. It can be trained on all of your policies and anything else you need to add to it. And also it's super easy for anyone else to use too. We actually have an example as you can see right here. So you can just custom train
it. You can edit the GPT. You can add custom instructions, upload knowledge files, etc. That's probably the easiest, simplest solution, right? If you want something more advanced, then you could probably give access to your team to
buzz or something like that. We've got a new tutorial inside the classroom of the something super simple, I would stick with custom GPT. And usually the simplest solution is the best. If you're wondering how to create a custom GPT, we
actually have a full tutorial. So if you're wondering how to create a custom way that I would approach it. Great question. Another question we got here was from Se about Pi. Now Pi is pretty
cool. It's kind of like um a kind of like an AI agent that you can custom train similar to, for example, OpenClaw or Hermes. In fact, OpenClaw was actually built on Pi. So, it's a pretty cool setup. We've actually got a full
tutorial on Pi if you want to learn how to to use it. And we've also got a tutorial on how to use Pi to build your own open claw, which I'll link to, too. But yeah, it's pretty powerful setup. It just takes a lot more work to set up
example, if you use something like Hermes, Hermes is a lot simpler and much more ready out of the box. But this as well looks like you could build your own custom UI or you could integrate it. If there's a enough demand, we might
actually add it to the agent OS. I just don't know many people that actually use that. But if you want to see me build it, feel free to direct message me inside the profit boardroom. Another great tip from Danish here about using
goal and he's given some examples here. So this is like a really nice in-depth tutorial. Thank you so much Danish for posting this. I've actually tested goal posting this. I've actually tested goal mode with codeex and also with Hermes
and we have forward slashgo built into the system. So for example in Hermes here you can see we've got goal mode and then also over in codeex on the agent OS we have goal mode as well. It's a really powerful feature and it can also code
out for days which is super exciting. We got another question from John which is did anthropic remove fable from claude code? They actually didn't. So you can see for example inside the Claude desktop here we have Fable 5. So Fable 5
is definitely still available inside Claude and I think they actually announced that it'll be permanently part of the Claude subscription. So you should still be able to get access and actually it's
much better than Opus. Opus 5 right now is just too slow to use. This pretty interesting. So John was sharing that according to open routters data, Hermes agent is ranked as the most used app in the platform which I've seen as well. Um
and three times more tokens are used with Hermes agent versus Claude which is insane when you think about it. I've tried both Hermes agent and Claude and we've got both built into the agent OS as well. For me personally, I still
prefer Claude simply because of the model Fable 5, but I understand why so many people are using Hermes agent instead because you can use free or local APIs directly inside Hermes. And also, you can customize it a lot more in
terms of the way you use it, how it's set up, how it looks, how you can build custom workflows like the Hermes Oracle that we've built. So, I could totally that we've built. So, I could totally understand why Hermes has grown so much.
We got a question here about Buzz. What is Buzz? So, Buzz is basically like Slack for AI agents. And what it means essentially, we've got a full tutorial inside the new daily update section, but what it means essentially is like you
what it means essentially is like you can have a AI powered version of Slack where you can add your agents, give them specialties. It can be used with subscription and also you can even operate it with Hermes. We've got a
tutorial on that too. So, it's kind of like a beginner friendly way to use AI agents that's fun and very visual. Muhammad said that first day in the community and he went all in. So, he recently joined the community and
started building out the agent OS which is pretty cool. And he actually by the is pretty cool. And he actually by the end had a fully automated SEO machine built specifically for his business. This is amazing. Congrats for day one
progress. Sounds like you absolutely crushed it and well done. Ed was asking about is anyone building like education technology with the agent OS and Hermes. The number one way that I can see people using this is the voice agent would be
great at for example teaching languages because you could go back and forth with learn a language for it and you could actually fine-tune or custom train it directly for teaching languages. I also think that the video manager inside the
agent OS would be great because you just give it research or you give it a topic voice over, everything else. And I think that'd be great for creating like course videos as well. So it can teach people
things in an automated way that's easy to understand and pretty fun as well. And for the voice agent, not just learning languages, but teaching people anything on a a call. So for example that could be learning marketing or
learning SEO and it's custom trained on all the stuff you know. Another question from Peter here who says first of all I want to say a big thank you to Julian for creating such an amazing agent OS provided the foundation which is amazing
and his goal is not to replace or rewrite the agent OS but just adapt it be fair and so these are the modules that he wants to use. So that sounds really cool. Like what I would do if you have custom ideas for the agent OS is
speak to the agent that actually set it up for you using the zip file and then from there explain exactly what you want from it. So for example, you've listed from it. So for example, you've listed some ideas above tell the agent exactly
what you want. Test it, iterate it, and then go from there. Whenever I'm building custom features into the agent OS, that's exactly what I'll do. So for example, we actually build custom features with Fable 5 and I was building
a feature earlier with Nine Root, right? Nine Root, it's a free model you can plug into the Agent OS. And when I actually tested it, it broke, but then I finally fixed. And if you follow that three-step formula of describing what
you want, testing it, and then giving more feedback to the agent, eventually that's the approach that I would recommend for you. And that is basically profit volumes who have answered every single question. Lots of great questions
question personally like this and I create a video tutorial for you. Feel air profit volume link in the comments description or go to the air full setup with all of this amazing stuff. So you can see how many powerful
the cool things you can do with it. You can get all of this inside the AI profit board community. go to the new daily update section and then go to the agent okay, doesn't it like cost a lot of tokens to use something like this? So,
we actually have a full training on exactly how to run an agent operating system for free. So, if you're worried about tokens, you can minimize them or a token minimization playbook here that shows you exactly how to reduce the
amount of tokens you use. So, you get all of this inside the air form. You week. So, you can jump on a live call, ask questions, get help and support. And in your local area who are building with AI agents like you. So feel free to get
it. Link in the comments description or just go to the aiprofitbomb.com.